Human Aging: (1)Functional magnetic resonance imaging (fMRI) detected areas of brain blood flow activation in human subjects who were presented visual texture patterns that differed in the density of features, such as contours and blocks. A linear relation between blood flow activation in the visual (striate) cortex and the density of the textures indicated that certain striate cortical neurons have an intrinsic ability to extract and process visual textures from an image. (2) For some 40 years, human brain blood flow and glucose metabolism have been considered to decline with age, suggesting reduced brain functional activity. We showed, however, that brain glucose metabolism, measured by positron emission tomography (PET) in the " resting state" and corrected for brain atrophy, did not decline significantly with age in healthy men and women. Thus, brain energy metabolism per gram actual brain is age-invariant in the absence of brain disease (in contrast to its being reduced with disease, see below). Alzheimer disease (AD): (1) AD patients with prominent visual disturbances (Balint's syndrome) showed glucose hypometabolism in brain visual areas on PET, before and after metabolism was corrected for brain atrophy. Pathological senile plaques and neurofibrillary tangles were found postmortem in the hypometabolic areas. Thus, unlike healthy aging, intrinsic brain metabolism even after atrophy correction is reduced in AD due to underlying neuropathology. (2) A multicenter longitudinal study, involving the Section on Brain Physiology and Metabolism at the NIA, was conducted on 284 suspected AD patients who underwent PET and cognitive testing. In neuropathologically confirmed cases, the PET patterns identified patients with AD or any neurodegenerative disease with a sensitivity of 94% and specificities of 73% and 78%, respectively. The likelihood of a progressive versus nonprogressive course following a single negative brain PET scan was 0.10. The initial PET scan predicted subsequent disease with a probability of 0.001. Thus, PET can be used for the early diagnosis of patients with AD or neurodegenerative disease generally. A negative scan makes the existence of disease highly unlikely. (3) Activation paradigms in which stimulus intensity is varied parametrically can be used to evaluate synaptic integrity in AD patients by means of PET or functional magnetic resonance imaging (fMRI). Activation scanning can be used to characterize the course of AD, to evaluate drug efficacy, to make an early diagnosis, and to identify genetically at-risk affected subjects prior to appearance of clinical dementia. (4) Brain activation in subjects who were presented a delayed match-to-sample visual stimulus test demonstrated different patterns in mildly-demented AD and control subjects, thus distinguishing the two groups. The abnormal pattern in the AD patients was associated with reduced accuracy as response delay was increased from 1 to 16 seconds, whereas accuracy was unchanged in the controls. The abnormal AD PET pattern suggested functional disconnection between prefrontal cortex and hippocampus. (5) The volume of the anterior corpus callosum, measured with magnetic resonance imaging (MRI), is a surrogate marker of brain metabolic dysfunction in AD. In AD patients, its size was reduced in proportion to reduced glucose metabolism, measured with PET, in frontal and parietal cortical regions. (6) In AD patients, 1H-magnetic resonance spectroscopy (MRS) with external standards demonstrated a reduced brain concentration of N-acetylaspartic acid but increased concentrations of myoinositol and creatine. As the latter increases were statistically significant in mildly demented AD patients, 1H-MRS can be used to diagnose early AD.